Kapanadze Cousin - DALLY FREE ENERGY

[NickZ]

  Itsu and All:
  So I've wound the new yoke, and installed it.  The fet ringing is slightly less but it's hard to tell because I'm running a 24v 18w car bulb on the yoke's input now. Both fets are ringing similar as they were before, maybe a little better, but I'd need to run on the PSU at 24v to really be able to compare.
   The 2nd image is the gate signals.
   The 3rd image is taken with the scope on volt/div on blank (50?), and the sec/div at 1us.
The expanded ringing scope shot is on .5us, and the same volts/div again on blank (50?). 
   Not sure what value is for the volt/div at the blank setting, so I figure that it's possibly at 50.
   Here's some pics, let me know what else you may need to figure the snubber values.
   
   

[itsu]

Nick,

looking better,

If the expanded ringing was with 0.5us setting, then the tops of the first 2 peaks are slightly longer then 1 division, so i made it 0.505us, see screenshot below.
This means 1.98Mhz ringing frequency (http://www.calctool.org/CALC/other/converters/freq)

Now follow the step 2 of this procedure:  https://www.maximintegrated.com/en/app-notes/index.mvp/id/3835   meaning you have to add a shunt capacitor
across the drain to ground to reduce the resonant frequency to approximately half or 990Khz (means a time of 1.01us between the first 2 peaks).
Use a good quality ceramic or film capacitor able to handle the peak voltage you see on your scope.

When you have found the correct cap value, take another screenshot of the drains (expanded) and show it, then we can continue with step 3 (Calculate the parasitic inductance).
   
Did you tune the TL494 for resonance of the wima cap / inductor?  What frequency was this resonance frequency?


Thanks Itsu

[NickZ]

  Itsu:
  I believe that the 3 division gate signal in the image that I posted was running at 16.7KHz.

  I've been looking to see what caps I have that might work for the snubbers.
But, it's been hard to find a matching pare of the right ones. Like a couple of 222, or so.
I'll continue looking tomorrow for some caps. Otherwise I'll have to order some, which can take a while.
  One question, when you say that the caps should go between the drain and the ground. Is that the same as between the drain and source? Same thing?

   I was looking at some of your Ruslan replication videos, and could see that you had a bit of a hard time with those "nasty spikes" also. Seams like most everyone has. But, you were able to tame them, it seams.
 
   If I touch the fet heatsink with my hand and with bare feet, I can reduce the spikes on the ringing signal by about 25%.  So, I'll be looking some more into the groundings, and such also. Looks like the proper earth grounding is very important for the TL494 board systems. Which seams to help a lot with reducing those spikes. There must be a way to keep the spike downs even without the snubbers, as Stalker doesn't use snubbers, and he's been at this for a while now, also. But maybe he should.
Akula and Ruslan used snubbers, on their fets.
   Anyway, I'll install some caps tomorrow and see what happens.
   

[TinselKoala]

Nick, the Source pins of the mosfets are connected to the circuit common negative aka "ground", right?

And the body tabs of the mosfets are internally connected to the Drains. So when you touch your heatsinks, and you have bare feet, you are providing a "snubber" capacitive-resistive connection from Drain to Ground through your own body.

At this point I am unsure of the exact circuit you are using, so it might be good if you could provide it again.


By the way, your focus problem on your scope is almost certainly due to one or more bad resistors in the Focus control's voltage divider, made of a string of 5 or six 1/2 watt resistors of 510k on the main board, resistors R888, R889, R890, R891, R892 and possibly the 75k 1/4 watt R894. These are all grouped together in one place on the board. This is a very easy fix as these things go.

[itsu]

Nick,

my question is if you are running your setup with the wima cap / inductor in resonance as that has a big influence on the spikes / ringing.


Anyway, yes i meant that the snubber test cap should go between drain and source (= ground).

Good point from TK, if your MOSFETs are not isolated from the heatsinks, touching the heatsinks will cause you to be a snubber!

With your 2Mhz ringing signal i guesstimate you should need a test cap of around 1.5nF to half it to 1Mhz

Itsu